Search Linux Wireless

Re: [PATCH RFC v2 2/8] bus: mhi: host: add new interfaces to handle MHI channels directly

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 





On 2/1/2024 6:00 PM, Manivannan Sadhasivam wrote:
On Wed, Jan 31, 2024 at 03:39:26PM +0800, Baochen Qiang wrote:


On 1/31/2024 2:19 AM, Manivannan Sadhasivam wrote:
On Mon, Nov 27, 2023 at 06:20:16PM +0200, Kalle Valo wrote:
From: Baochen Qiang <quic_bqiang@xxxxxxxxxxx>

When using mhi_power_down_no_destroy() MHI hosts need to unprepare MHI channels
by themselves.  Similarly, MHI stack will also not create new MHI device since
old devices were not destroyed, so MHI hosts need to prepare channels as well.
Hence add these two interfaces to make that possible.

Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30

Signed-off-by: Baochen Qiang <quic_bqiang@xxxxxxxxxxx>
Signed-off-by: Kalle Valo <quic_kvalo@xxxxxxxxxxx>
---
   drivers/bus/mhi/host/main.c | 107 ++++++++++++++++++++++++++++++++++++
   include/linux/mhi.h         |  20 ++++++-
   2 files changed, 126 insertions(+), 1 deletion(-)

diff --git a/drivers/bus/mhi/host/main.c b/drivers/bus/mhi/host/main.c
index d80975f4bba8..3f677fc628ad 100644
--- a/drivers/bus/mhi/host/main.c
+++ b/drivers/bus/mhi/host/main.c
@@ -1669,6 +1669,58 @@ int mhi_prepare_for_transfer_autoqueue(struct mhi_device *mhi_dev)
   }
   EXPORT_SYMBOL_GPL(mhi_prepare_for_transfer_autoqueue);
+static int ____mhi_prepare_for_transfer(struct device *dev, void *data)

"__mhi_prepare_all_for_transfer"

This is to prepare one single child device, I don't think a name like
__mhi_prepare_all_for_transfer (with 'all' inside) make sense, right?
How about changing to "mhi_prepare_dev_for_transfer" or
"mhi_prepare_single_for_transfer"?


I think most of the checks in this function can be moved inside
mhi_prepare_for_transfer() API. With that you can just reuse the API without
adding a new helper.

For autoqueue channels, you can add another API
mhi_prepare_all_for_transfer_autoqueue() just like
mhi_prepare_for_transfer_autoqueue() to maintain uniformity.

- Mani
If we do that, we need to call two APIs together, does it make sense? From the view of an MHI user, what we want is an API to prepare all channels, no matter whether a channel is configured as autoqueue or non-autoqueue, we don't care about it.

And besides, I don't think there is a scenario where we need to use them separately. So if we always need to use them together, why not merge them in a single API?



+{
+	struct mhi_device *mhi_dev;
+	struct mhi_chan *ul_chan, *dl_chan;
+	enum mhi_ee_type ee = MHI_EE_MAX;

Reverse Xmas order, please.

+
+	if (dev->bus != &mhi_bus_type)
+		return 0;
+
+	mhi_dev = to_mhi_device(dev);
+
+	/* Only prepare virtual devices that are attached to bus */

"Only prepare virtual devices for the channels". Here and below.

+	if (mhi_dev->dev_type == MHI_DEVICE_CONTROLLER)
+		return 0;
+
+	/* There are cases where there is no MHI client driver matches
+	 * this device, we are not allowed to do prepare for it.
+	 */

Use the preferred style for comment:

	/*
	 * ...
	 */

+	if (!mhi_dev->id)
+		return 0;
+
+	ul_chan = mhi_dev->ul_chan;
+	dl_chan = mhi_dev->dl_chan;
+
+	/*
+	 * If execution environment is specified, remove only those devices that
+	 * started in them based on ee_mask for the channels as we move on to a
+	 * different execution environment
+	 */
+	if (data)
+		ee = *(enum mhi_ee_type *)data;
+
+	if (ul_chan && ee != MHI_EE_MAX && !(ul_chan->ee_mask & BIT(ee)))
+		return 0;
+
+

Remove extra newline.

+	if (dl_chan && ee != MHI_EE_MAX && !(dl_chan->ee_mask & BIT(ee)))
+		return 0;
+
+	if (dl_chan->pre_alloc)
+		return mhi_prepare_for_transfer_autoqueue(mhi_dev);
+	else
+		return mhi_prepare_for_transfer(mhi_dev);
+}
+
+int mhi_prepare_all_for_transfer(struct mhi_controller *mhi_cntrl)
+{
+	return device_for_each_child(&mhi_cntrl->mhi_dev->dev, NULL,
+				     ____mhi_prepare_for_transfer);
+}
+EXPORT_SYMBOL_GPL(mhi_prepare_all_for_transfer);
+
   void mhi_unprepare_from_transfer(struct mhi_device *mhi_dev)
   {
   	struct mhi_controller *mhi_cntrl = mhi_dev->mhi_cntrl;
@@ -1684,3 +1736,58 @@ void mhi_unprepare_from_transfer(struct mhi_device *mhi_dev)
   	}
   }
   EXPORT_SYMBOL_GPL(mhi_unprepare_from_transfer);
+
+static int ____mhi_unprepare_from_transfer(struct device *dev, void *data)

__mhi_unprepare_all_from_transfer

same as above.


+{
+	struct mhi_device *mhi_dev;
+	struct mhi_chan *ul_chan, *dl_chan;
+	enum mhi_ee_type ee = MHI_EE_MAX;
+
+	if (dev->bus != &mhi_bus_type)
+		return 0;
+
+	mhi_dev = to_mhi_device(dev);
+
+	/* Only unprepare virtual devices that are attached to bus */
+	if (mhi_dev->dev_type == MHI_DEVICE_CONTROLLER)
+		return 0;
+
+	/* There are cases where there is no MHI client driver matches
+	 * this device, so it is not probed or prepared, no need to
+	 * do unprepare for it.
+	 */
+	if (!mhi_dev->id)
+		return 0;
+
+	ul_chan = mhi_dev->ul_chan;
+	dl_chan = mhi_dev->dl_chan;
+
+	/*
+	 * If execution environment is specified, remove only those devices that
+	 * started in them based on ee_mask for the channels as we move on to a
+	 * different execution environment
+	 */
+	if (data)
+		ee = *(enum mhi_ee_type *)data;
+
+	if (ul_chan) {
+		if (ee != MHI_EE_MAX && !(ul_chan->ee_mask & BIT(ee)))
+			return 0;
+	}
+
+	if (dl_chan) {
+		if (ee != MHI_EE_MAX && !(dl_chan->ee_mask & BIT(ee)))
+			return 0;
+	}
+
+	mhi_unprepare_from_transfer(mhi_dev);
+
+	return 0;
+}
+
+int mhi_unprepare_all_from_transfer(struct mhi_controller *mhi_cntrl)
+{
+	return device_for_each_child(&mhi_cntrl->mhi_dev->dev, NULL,
+				     ____mhi_unprepare_from_transfer);
+}
+EXPORT_SYMBOL_GPL(mhi_unprepare_all_from_transfer);
diff --git a/include/linux/mhi.h b/include/linux/mhi.h
index ae092bc8b97e..dcf62a57056a 100644
--- a/include/linux/mhi.h
+++ b/include/linux/mhi.h
@@ -668,7 +668,7 @@ static inline void mhi_power_down(struct mhi_controller *mhi_cntrl, bool gracefu
    * destroy struct devices. This is a variant for mhi_power_down() and is a
    * workaround to make it possible to use mhi_power_up() in a resume
    * handler. When using this variant the caller must also call
- * mhi_prepare_all_for_transfer_autoqueue() and
+ * mhi_prepare_all_for_transfer() and

This change belongs to previous patch.

    * mhi_unprepare_all_from_transfer().
    *
    * @mhi_cntrl: MHI controller
@@ -842,4 +842,22 @@ int mhi_queue_skb(struct mhi_device *mhi_dev, enum dma_data_direction dir,
    */
   bool mhi_queue_is_full(struct mhi_device *mhi_dev, enum dma_data_direction dir);
+/**
+ * mhi_prepare_all_for_transfer - if you are using
+ * mhi_power_down_no_destroy() variant this needs to be called after
+ * calling mhi_power_up().

Add info about what this API does also.

+ *
+ * @mhi_cntrl: MHI controller
+ */
+int mhi_prepare_all_for_transfer(struct mhi_controller *mhi_cntrl);
+
+/**
+ * mhi_unprepare_all_from_transfer - if you are using
+ * mhi_power_down_no_destroy() variant this function needs to be called
+ * before calling mhi_power_down_no_destroy().

Same as above.

- Mani

+ *
+ * @mhi_cntrl: MHI controller
+ */
+int mhi_unprepare_all_from_transfer(struct mhi_controller *mhi_cntrl);
+
   #endif /* _MHI_H_ */
--
2.39.2








[Index of Archives]     [Linux Host AP]     [ATH6KL]     [Linux Wireless Personal Area Network]     [Linux Bluetooth]     [Wireless Regulations]     [Linux Netdev]     [Kernel Newbies]     [Linux Kernel]     [IDE]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite Hiking]     [MIPS Linux]     [ARM Linux]     [Linux RAID]

  Powered by Linux